Acquired aphasia is a lifelong disability that affects functional communication and consequently, quality of life. We will test the efficacy of a treatment to improve sentence production impairment in nonfluent, agrammatic aphasia, which is characterized by a reliance on content words, paucity of function words, and use of simpler sentence structures. The impairment has been attributed in part to reduced processing capacity which makes it difficult to access syntactic structure during sentence production or comprehension. The treatment method is based on syntactic (or structural) priming, which uses repetition of a sentence with a particular syntactic structure (e.g., passive transitive) to facilitate access to that same structure in subsequent picture description trials that could be described with that same structure. Residual activation of the priming sentence makes it more accessible for a short period, and thereby increases the probability of its being retrieved to describe a pictured scene. This treatment approach can be used regardless of the severity of the sentence production impairment. This is especially noteworthy because there are few treatments for more severe agrammatic aphasia. In this project, we will examine several ways to enhance syntactic priming effects on sentence production in persons with agrammatic aphasia, further increasing its value as a treatment method for this population. First, we will vary semantic and syntactic argument structure of the treatment stimuli in ways that reduce the number of semantic elements to be syntactically encoded but without altering the surface form of the sentence. In Study 1, we aim to show that verb-particle sentences with three semantic elements (The driver is turning off the lights) show more robust priming effects than sentences with prepositional phrases and four semantic elements (The driver is turning off the road). In Study 2, the coherence of verb-object phrases in transitive will be varied to show that transitive with more coherent verb- object phrases (mowing the lawn) show more robust priming effects than those with less coherent verb-object phrases (cooking the soup). We will test the related hypothesis that transitive with coherent verb-object phrases are more effective primes of those without coherent verb-object phrases than vice versa because they are more likely to result in successful production of the priming sentence. In Study 3, we will manipulate the priming procedure using an augmentative communication (AAC) device that allows more time to explicitly process the priming sentence. Potential outcomes of these studies will inform theories of the relationship of short-term memory and sentence processing, providing evidence that agrammatic sentence production is related to reduced processing capacity. They will provide additional evidence that syntactic priming engages implicit learning processes and that the addition of explicit learning procedures enhances the effect. Clinically, these studies will provide evidence that syntactic priming treatments effectively improve sentence production in agrammatic aphasia, even in more severe cases when production is limited to 1-3 word utterances.